Google Maps Hackers Just Don’t Get It
[on my high horse]
I wrote up a long rant about how ignorant these Google Maps hackers were to think that one should obfuscate history on old imagery, but I decided to delete that post because I don’t think they can handle it.
[/on my high horse]
Look, stick to making great hacks and leave the back end GIS source to professionals. There isn’t anything wrong with the images that MSN Virtual Earth uses. Sure they are old, but a great GIS database includes time and historic aerial/satellite imagery is wonderful to see changes in places. If there isn’t a better reason to keep investing in GIS professionals that post on Google Maps Mania, I can’t think of one.
Rather than tune out Jack Dangermond at Where 2.0, you guys need to listen to what he is saying about this stuff because you might learn a thing or two about GIS. The power of products such as Google Maps is in more than just an easy to use API. It’s the data behind the server and without GIS professionals, you’ll end up with what happened at MSN Virtual Earth on launch (no metadata available to tell users the data was old) and Google Maps hackers (wanting to reduce the accuracy of datasets).
Update - Let me just add that Google Maps doesn’t have metadata attached to their imagery either. The copyright says 2005, but that doesn’t mean that the images date from 2005. At least were I live they date from about late 2002/early 2003. Google too should let users know how old the imagery is.
It’s all moot. Hardly anyone, including the GIS professionals, attaches metadata anyways. Even when you do, nobody else knows what do with it.
@Swicths, and James,
The question that I have for ‘professionals’ who perpetually harp on Google Earth is:
What amount of accuracy are you looking for, when the imagery database has been enhanced for visualization practice within the viewer?
To include metadata as a reference is one thing - but to what value that metadata becomes when the imagery is being deployed within a viewer [with radical enhancements] - is the moot position. And in my mind, entirely knit-picky.
The beauty of Google Earth and now, Google Maps, is its provisioning of allowing anyone - a scientist, a professional, and even ameteurs interested in GIS sciences, to produce datasets that can be viewed in these environments.
What this suggests then, is a lack or unwillingness on the part of anyone who is producing datasets to not include reference metadata as an aside, or linked by annotation within the viewer model.
What about this issue is really all that confusing, aside from a handful of people who fell in love with a technology that’s dying because of others’ popularity?
The world does inevitably change, if I’m not mistaken. Either embrace it, or be left out.
If what you’re trying to say though - which I think is the case - is that you’d prefer that metadata for each image tile to be present or linked, so that it can be referenced within the viewer - then yes, I’ll agree with you that it would be value-added for certain people. But the practicality of referencing all the imagery in many ways is moot in relation to the editing that has taken place for the default database.
If you wish to include metadata reference in Google Earth, then I suggest outputting a polyline shape grid with each cell being annotated and linked to each tile’s metadata - and allow that to be pulled in as your reference layer. Not very complicated, is it?
I mean, unless everyone just wants to be able to download some ‘free’ imagery into their own apps - like is possible with TerraServer. But I don’t think the imagery providers for Google Earth would entirely appreciate that.
And sorry for the rant. Just call me one of those whacky ‘professional’ purists, when it comes to metadata.